30results about How to "Avoid vanishing gradients" patented technology

The invention provides a fan blade crack detection method, which comprises the following steps of: for a fan blade image acquired by an unmanned aerial vehicle, performing deblurring processing on the image through a pre-processing algorithm designed by the invention, and overcoming the problem of motion blurring when the unmanned aerial vehicle acquires the fan blade image; and carrying out crack detection on the acquired image by adopting a crack detection neural network model improved based on YOLOv8. According to the crack detection model, self texture interference of the blade can be avoided, the crack detection precision is improved, crack characteristics of different scales are captured through a Stem module of a multi-branch structure, a CAM module is added into a C2f-1 module, crack attention is integrated, and crack related characteristics are enhanced; a residual connection mode of a residual block is designed in a C2f-2 module to avoid the problem of deep network gradient disappearance, gradient flow and effective transmission of crack characteristics are guaranteed, response of crack-related channels is enhanced through channel re-calibration, and irrelevant channels such as blade textures are inhibited.

The application provides a method for constructing a regulation network of a potato JA pathway in response to Cd stress. The method relates to the technical field of data processing, and comprises the following steps: obtaining multi-omics time series data of potatoes under the conditions of cadmium stress and methyl jasmonate treatment; performing signal decomposition and feature purification on the multi-omics time series data by using a weak stationary signal decomposition method based on sequence patterns to obtain a plurality of intrinsic mode components; constructing a dynamic regulation network model to capture multi-scale change patterns of response characteristics in the time dimension and the concentration dimension, thereby extracting deep features; inputting the intrinsic mode components into the trained dynamic regulation network model to output an attention weight matrix, thereby forming a regulation network representing the dynamic regulation intensity of the jasmonic acid pathway; and solving the optimal exogenous methyl jasmonate concentration for promoting biomass and inhibiting cadmium accumulation by optimizing an objective function. The application provides a high-precision model tool for analyzing the regulation mechanism of plant hormone signals under heavy metal stress.

The invention discloses a deep counterfeiting detection method based on double-flow fusion and adaptive feature enhancement, and belongs to the technical field of computer vision and digital media security. According to the method, a double-flow feature extraction network is constructed, a high-low feature adaptive enhancement module (HLFAE) is adopted in a spatial flow to decompose multi-scale texture features, and micro texture expression of a forged area is enhanced in combination with expansion convolution and a channel attention mechanism; introducing a high-frequency sub-band capable of learning a discrete wavelet transform self-adaptive decomposition image into a frequency domain flow, and amplifying frequency domain artifact features through a convolutional network; a multi-modal enhanced feature attention module is designed to dynamically fuse space and frequency domain features, significant features are weighted through a multi-scale convolution kernel and a double attention mechanism, and feature interaction consistency is improved based on a cross-modal contrast enhancement (CMCE) module. According to the method, the highest AUC value of 99.63% is achieved on data sets such as FaceForce + + and Celeb-DFv2, the cross-domain generalization ability and the anti-disturbance robustness are remarkably improved, and the method is suitable for financial risk control and media content auditing.

The invention discloses a photovoltaic MPPT (Maximum Power Point Tracking) control algorithm, which relates to the technical field of photovoltaic control algorithms, utilizes the characteristics of simple structure, high convergence speed, high global optimization capability, wide applicability and the like of a gradient optimization algorithm, and improves the defects of premature convergence and sensitivity to control parameters of the gradient optimization algorithm. Meanwhile, the photovoltaic output energy efficiency and the stability of the photovoltaic output energy efficiency serve as the evaluation basis, an MPPT algorithm performance evaluation model is constructed through an analytic hierarchy process, photovoltaic MPPT performance is evaluated, and GTO performance is verified and improved.

This invention discloses an implicit reconstruction method for point cloud surfaces based on a piecewise learning strategy. The invention employs a piecewise surface representation and trains an implicit reconstruction network for point cloud surfaces targeting the local symbolic distance field of a 3D shape. The method includes the following steps: using discrete point cloud data as input, a farthest-point sampling strategy is used to generate initial patches; the offset of each sampling point within each patch relative to the patch center is calculated as the relative position of the sampling point. The latent features of the patch are obtained in the neural network encoder, and the symbolic distance values ​​of the relative positions of each sampling point are obtained in the neural network decoder. The relative positions of sampling points located in overlapping areas of different patches are weighted and summed to obtain their corresponding symbolic distance values. A mesh model of the 3D shape is obtained using the Marching Cube algorithm. This invention can reconstruct the overall shape of an object while preserving the fine details of the original shape, and it is robust to point cloud normals and noise.

This design is a multi-task automatic EEG detection and prediction system for epilepsy. The system first receives the EEG signals from epilepsy patients to be identified or the collected patient EEG signals used for training the module through a preprocessing module. The preprocessing module performs noise reduction, filtering, segmentation, and normalization on the signals, and outputs the processed patient EEG signals to the training module. In the training module, the collected patient EEG signals are used for training and testing. First, feature extraction is performed using a multi-scale convolutional network. Then, the detection and prediction branches incorporating a dual attention mechanism are trained, and the trained model parameters are saved for use in the detection and prediction module. Finally, the detection and prediction module provides detection and prediction results for the EEG signals from epilepsy patients to be identified. This invention employs a modular design, achieving real-time detection and prediction of noisy, unbalanced EEG signals, providing a reliable system for practical applications.

The invention discloses a wood defect detection method based on an improved YOLOv11 model. The method comprises the following steps: acquiring a surface image of wood to be detected; the to-be-detected wood surface image is input into an improved YOLOv11 model, a detection result is obtained, the improved YOLOv11 model completes feature extraction and fusion through a backbone network module, a neck module and a detection head module in sequence, and before the detection head module, the detection head module completes feature extraction and fusion of the to-be-detected wood surface image; a rectangular attention module, a texture removal attention module and a circular coordinate attention module are introduced in parallel, and reweighting is carried out on the feature maps of the corresponding hierarchies; and respectively sending the reweighted three layers of feature maps into the Detect detection heads in the corresponding detection heads, and outputting defect category and position information.

The invention provides a microseism event identification method and system based on multi-dimensional feature fusion ensemble learning, and the method comprises the steps: collecting and processing microseism data, and constructing a multi-dimensional feature tensor; constructing a multi-dimensional feature fusion ensemble learning network model by using a feature fusion subnet and a target fusion subnet based on the multi-dimensional feature tensor; and based on the multi-dimensional feature fusion ensemble learning network model, identifying target micro-seismic data to obtain a micro-seismic event identification result. According to the method, efficient and accurate microseism event identification is realized through collaborative operation of data preprocessing and the secondary subnet, and the application value is remarkable.

This invention discloses a dynamically densely connected spatiotemporal feature decoupling network for recognizing cross-view gait, relating to the field of computer vision. It includes: an initial feature processing module, a dynamically dense spatiotemporal decoupling feature extraction module, and a feature enhancement processing module. The module uses dense spatiotemporal feature decoupling blocks and concatenation operations to achieve the sharing of shallow and deep network features, thereby addressing the problem of insufficient representation ability. Simultaneously, it employs an enhanced convolutional block attention mechanism to allow the network to focus on more important gait features. Finally, the feature enhancement processing module processes the five-dimensional feature mapping into multiple lateral features and performs batch standardization of the features to enhance representation ability and model generalization ability. This allows for the mining of correlations between shallow and deep features, alleviating the problem of insufficient information representation ability.

The application discloses a remote sensing image landslide detection method and system, and belongs to the technical field of landslide detection. The method adopts a data enhancement method to expand a landslide data set, and solves the problem of a small number of landslide samples in a training set. DarkNet53 is used to replace a commonly used backbone feature extraction network of FasterR-CNN, so that the problem of gradient disappearance caused by a too deep network is avoided, and the detection precision is improved.

The invention discloses a remote sensing sea fog detection method and system based on deep learning, and the method comprises the following steps: obtaining an original multiband remote sensing image for sea fog detection, and carrying out the standardized data preprocessing; the preprocessed multiband remote sensing image data are input into a preset deep learning network model, a fused feature map is output, and the deep learning network model comprises an encoder and a decoder which are combined in a jump connection mode; and processing the fused feature map through a SoftMax activation function to generate a pixel-level sea fog probability map, and completing image segmentation of the sea fog boundary. Compared with the prior art, the sea fog detection method can overcome the problems of fuzzy sea fog boundary, insufficient multi-scale feature extraction, shallow space information loss and the like of the sea fog detection method of the existing standard deep learning model.

The invention provides an article detection method based on feature enhancement and parallel channel attention. The method comprises the following steps: acquiring a millimeter wave image to be detected; sending the millimeter wave image into a filtering module, and carrying out regional feature enhancement and channel adjustment to obtain a feature-enhanced image; sending the image after feature enhancement to a backbone network, and performing feature extraction through dense residual blocks and jump connection to obtain a feature map; a parallel channel separation attention module is introduced between the backbone network and the feature pyramid network, decoupling feature enhancement of space and channel dimensions is carried out on the feature map, and an optimized fusion feature map is obtained; the optimized fusion feature map is sent to a feature pyramid network for feature aggregation, and aggregation feature information is obtained; and carrying out classification and bounding box positioning on articles with specific sizes by using the aggregation feature information through a detection output head to obtain a detection result. The method is used for solving the problem of inaccurate detection of objects with specific sizes.

This application discloses a training method and related apparatus for a quantum generative adversarial network (GAN), belonging to the field of quantum computing technology. The GAN includes a generator and a discriminator. The method includes: using the generator to obtain generated samples against random noise; using the discriminator to distinguish between real samples and the generated samples, obtaining a discrimination result; updating the parameters of the generator and the discriminator based on the discrimination result, the loss function of the generator, and the loss function of the discriminator, to obtain a trained GAN. At least one of the generator and the discriminator includes a quantum convolutional layer and a quantum residual neural module connected sequentially. The quantum residual neural module includes a first qubit and a second qubit for encoding and evolving each feature data. Two qubits; a first quantum logic gate and a second quantum logic gate acting on the first qubit for performing an identity mapping operation on feature data; a third quantum logic gate acting on the second qubit and entangled with the first qubit between the first and second quantum logic gates, wherein the third quantum logic gate includes an encoding module for encoding feature data and parameterized training logic gates located before and after the encoding module for implementing residuals, and the encoding module includes an encoding logic gate and entanglement gates located before and after the encoding logic gate, the entanglement gates realizing the entanglement of the first and second qubits; the second qubit corresponding to one feature data at an adjacent position is the first qubit corresponding to another feature data, the feature data being obtained based on a quantum convolutional layer. Applying this application can effectively suppress the gradient vanishing problem and reduce the waste of training resources caused by training failures.

The invention relates to the technical field of saline-alkali land improvement, in particular to a saline-alkali land improvement parameter optimization system based on a multilayer deep neural network, comprising a data acquisition module used for acquiring soil salinity data, soil moisture data, soil nutrient data and soil pH value data of target saline-alkali land, the data acquisition module is used for acquiring and forming time sequence soil parameter original data; and the data preprocessing module is used for carrying out missing value filling, abnormal value elimination and normalization processing on the original data of the time sequence soil parameters to generate a time sequence soil parameter data set. The multi-dimensional time sequence soil parameters of the target saline-alkali soil are acquired in real time through the data acquisition module, the data quality is optimized through the preprocessing module, and the long-term dependency relationship of the soil parameters along with the time change is effectively captured by using the multi-layer deep neural network module containing the long-short-term memory network layer or the gating circulation unit layer. And the problem of gradient disappearance or information attenuation of the existing model is avoided.

This invention belongs to the technical field of bearing fault diagnosis. To address the inaccuracy of existing bearing fault diagnosis methods, this invention proposes a fault diagnosis method and system for press bearings based on conditional generative adversarial networks (GANs). A generator is trained using operating condition vectors and Gaussian noise. A discriminator is trained using the health state features generated by the generator and the actual health state features. The generator and discriminator are then subjected to adversarial training to prevent the discriminator from distinguishing between the health state features generated by the generator and the actual health state features, as well as from distinguishing the matching between the health state features generated by the generator and the corresponding operating condition vectors. This results in a well-trained health state model. The health state model is then used to generate a dynamic health baseline for the press shaft to be diagnosed, thereby obtaining the fault diagnosis result for the press shaft and achieving high-precision fault diagnosis and early warning.

The present application belongs to the technical field of underwater acoustic imaging and three-dimensional reconstruction, and particularly relates to a three-dimensional reconstruction method based on forward scanning sonar images, which obtains two-dimensional gray scale images collected by a sonar, uses semantic segmentation to analyze the features of the images, obtains light and shade feature channels of highlight areas and shadow areas, and extracts boundary features; extracts gray scale gradients and gray scale ratios in the highlight areas, solves obstacle height profile points by nonlinear fitting and taking obstacle front edge points and shadow boundary points as constraints, and finally realizes three-dimensional topography reconstruction by fusing multi-array features. The present application can effectively suppress underwater artifacts and noise interference, realize high-precision three-dimensional reconstruction, has the advantages of lightweight model, low power consumption, etc., and is suitable for covert detection in complex underwater environments and underwater surveying and mapping scenes. The present application also provides a non-transitory readable recording medium storing a program of the method and a system containing the medium, and the program can be called by a processing circuit to execute the above method.

The present disclosure relates to a power distribution network service hot migration method and device, electronic equipment, storage medium and computer program product. The method comprises: obtaining resource information of each edge node in a plurality of edge nodes contained in a power distribution network and load information of services on each edge node; inputting the resource information of each edge node and the load information of the services on each edge node into a residual network contained in a trained service migration prediction model to obtain a feature representation; inputting the feature representation into a migration strategy decision module contained in the service migration prediction model to obtain a service migration instruction, wherein the service migration instruction is used to indicate a to-be-migrated service and a migration path and a migration destination edge node corresponding to the to-be-migrated service; and issuing the service migration instruction to an edge node containing the to-be-migrated service in the plurality of edge nodes, so that the edge node containing the to-be-migrated service performs hot migration on the to-be-migrated service based on the service migration instruction.

The application discloses a kind of offshore wind turbine fault identification method and system, storage medium, computer equipment, the method includes: obtaining the historical vibration time series signal of offshore wind turbine gearbox multiple operating states and pre-processing training sample, on the basis of traditional one-dimensional convolutional neural network, integration multi-scale convolution module, SE attention mechanism and residual network structure, construct new network model.Training time, model receives sample by input layer, middle layer extracts feature, weighting, output layer generates probability distribution and outputs state, after reaching standard, it predicts the operating state under the new signal of real-time acquisition.By multi-scale convolution, the features of different time scales are extracted, the adaptive weighting of feature channels is realized by combining the SE attention mechanism module, and the stability of deep network training is enhanced by using the residual structure, thereby realizing the accurate identification of complex vibration signals, effectively improving the diagnostic accuracy and robustness, and providing a scientific basis for the operation and maintenance of offshore wind turbines.

The invention discloses a down feather recognition method and system based on deep learning, and the method comprises the steps: S1, obtaining an original down feather image, carrying out the preprocessing of the obtained original down feather image, and obtaining a processed down feather feature map; s2, the obtained feature map is input into an improved residual network WT-ResNet model, the WT-ResNet model at least comprises a WT-ResNet layer, and the WT-ResNet layer processes the input feature map and then outputs a final feature map containing deep semantic information; and S3, inputting the output final feature map into a classification activation layer, obtaining a probability value of down feather image classification through activation function processing, and judging whether the down feather image is fresh down feather or recycled down feather based on the probability value.

This invention discloses a highly transferable adversarial example generation method and its application in a secure test range. The method applies principal component protected feature perturbation transformation to the intermediate layer feature map of a momentum-based iterative fast gradient symbolic attack framework. It identifies the principal and secondary components of the feature space using singular value decomposition (SVD), quantifies the redundancy between feature channels using Pearson correlation coefficient analysis, and achieves precise feature perturbation using an adaptive mask generation mechanism based on Bernoulli sampling and energy conservation constraints. This method can be embedded as a pluggable module into existing iterative gradient attack processes without modifying the main structure of the basic attack algorithm or retraining the source model. By dynamically registering hook functions during forward propagation to intercept and transform intermediate layer features, it significantly improves the transferability of adversarial examples on unknown target models. The method's high transferability and ease of use in the test range environment are verified, providing users with an effective tool for evaluating model robustness.

This invention discloses a multi-level land resource segmentation migration fine-tuning method with input alignment, comprising the following steps: acquisition and processing of remote sensing image data and label data related to the land resource segmentation task; selection and slicing of sampling areas based on prior knowledge of the study area; data augmentation to expand the dataset and pixel value normalization; partitioning the model dataset using stratified sampling techniques; determining the optimal initial learning rate through hyperparameter optimization, entering input feature alignment, and establishing a quick connection between the input alignment module and the model output; employing three Transformer architectures and their pre-trained models for land resource segmentation, and optimizing the model training process using gradient-value-constrained gradient clipping and two-stage training strategies; step six: segmentation prediction and visualization. This invention improves the model's training efficiency and segmentation performance, achieving high-precision multi-level land resource segmentation and recognition.

The application discloses a soft-hard interbedded rock mechanical parameter prediction method and system based on a residual attention network, relates to the technical field of rock mechanical parameter prediction, and has the advantages that the traditional neural network is prone to gradient disappearance when processing a deep network, which influences the training effect; the existing method lacks an attention mechanism and cannot effectively identify and strengthen key features; and the modeling capability for interlayer interaction is insufficient; the application provides a soft-hard interbedded rock mechanical parameter prediction method based on a residual attention network, which comprises the following steps: obtaining structure parameters and target mechanical parameters of a soft-hard interbedded rock sample; converting the structure parameters into an enhanced feature vector; constructing a residual attention network model; inputting the enhanced feature vector into the residual attention network model; and outputting a mechanical parameter prediction result and reliability evaluation information.

This invention discloses a recurrent neural network training method and system based on Ising machine optimization, belonging to the field of neural network training technology. The method transforms the weight training task of a recurrent neural network into a quadratic unconstrained binary optimization problem, mapped to an Ising model; it utilizes the physical evolution or annealing process of the Ising machine to obtain the spin configuration of the minimum energy state; finally, it decodes and restores the optimized weights and loads them into the recurrent neural network to complete the training. This invention achieves accurate representation of continuous weights through binary discrete encoding and offset matrices, supporting accelerated solutions using various Ising machine hardware (such as optical Ising machines, quantum annealing machines, etc.). Compared with existing gradient-based training methods, this invention completely avoids the gradient vanishing and gradient exploding problems, significantly improving the stability and accuracy of long sequence predictions. Simultaneously, based on advanced Ising machine computing equipment, the core training time can be shortened from seconds to milliseconds, providing a new technical path for efficient, low-power AI systems.

This invention discloses a bearing fault diagnosis method and system based on cross-modal feature alignment, belonging to the field of rotating machinery condition monitoring and fault diagnosis technology. The method includes: constructing sound feature vectors and vibration feature vectors based on the envelope spectrum of sound signals and the envelope spectrum of vibration signals to form a sound-vibration synchronization dataset; training the constructed cross-modal feature alignment diagnostic model using the training and validation sets in the sound-vibration synchronization dataset to obtain a trained cross-modal feature alignment diagnostic model, thereby obtaining a solidified cross-modal feature alignment diagnostic model; using the sound feature vectors from the test set in the sound-vibration synchronization dataset or the sound feature vectors to be diagnosed as input to the solidified cross-modal feature alignment diagnostic model to obtain the bearing diagnosis result. This invention can realize fault diagnosis under the same operating conditions based on sound vibration components, and further realize fault diagnosis across operating conditions.

This invention relates to the field of oil and gas reservoir exploration technology, specifically a method, apparatus, electronic device, and storage medium for predicting the content of multiple minerals. The method includes acquiring input features of the well section to be predicted; inputting these features into a multi-component mineral content prediction model; and outputting the corresponding prediction results for the content of each mineral component. The multi-component mineral content prediction model is trained using multiple samples on a pre-set deep learning model, which employs a strategy of bidirectional multi-scale feature extraction and cross-scale attention fusion. The model constructed by this invention can not only efficiently capture the global long-term trend and local abrupt fluctuations of logging curves, but also automatically learn the mutual constraints between minerals, achieving accurate mapping between multi-scale logging features and specific mineral categories. Therefore, the multi-component mineral content prediction model enables efficient and collaborative prediction of multiple minerals, providing an effective data foundation for oil and gas exploration and development, reservoir evaluation, and production capacity prediction.

The application discloses a kind of based on sub-domain self-adaptive multi-channel synthetic aperture radar moving target detection method, obtains the simulation and measured data of multi-channel synthetic aperture radar, makes network training and verification dataset, adds label for clutter and moving target;Subdomain self-adaptive residual network for ground moving target detection is built;The network training and verification data are preprocessed, input into subdomain self-adaptive residual network, save the optimal network after training ends;The radar data to be measured is slid window, and network test dataset is made;The preprocessed network test data are input into the saved network, and the predicted label is obtained;All prediction results are windowed, and the ground moving target detection result is obtained.The application trains network by using labeled simulation data and unlabeled measured data, reduces the difference between simulation and measured data by subdomain self-adaptive, solves the problem of lack of labeled data in the field of radar, and effectively improves the detection performance of ground moving target.

The invention belongs to the technical field of elastic backing plate appearance defect detection, and particularly relates to a railway fastener elastic backing plate defect detection method based on improved YOLOv8. The method comprises the following steps: S1, data processing; s2, constructing a model; s3, model training; s4, defect detection; the model is improved by taking YOLOv8x as a basic algorithm, an improved YOLOv8 model is constructed, and the improvement comprises the step of replacing all C2f modules in the YOLOv8 with a GELAN module; in a top-down path of the Neck part, an improved SAFM module is embedded before up-sampling operation of each layer; and a normalized Wasserstein distance NWD loss function is adopted as a bounding box regression loss function. Compared with a conventional YOLO algorithm, the detection method provided by the invention has the advantages that the detection precision is remarkably improved, and intelligent and automatic online detection is realized.

This invention provides a graph representation learning method based on a variational hypergraph mask autoencoder. The method includes: first, adaptively masking the input graph data according to node degree to generate a masked node feature matrix; then, obtaining initial node representations using an initial graph encoder incorporating an edge feature attention mechanism; subsequently, constructing a hypergraph structure based on the initial representations, and performing bidirectional information aggregation between nodes and hyperedges through hypergraph convolutional layers to output a hypergraph encoded representation; next, inputting the hypergraph encoding into a variational encoder to generate a probabilistic latent representation; finally, reconstructing node features using a decoder, calculating the total loss including reconstruction loss, KL divergence loss, and structure preservation loss, and optimizing the model using a progressive pre-training strategy. This invention effectively captures high-order association information of graph data, solves the problems of coarse masking and loss of structural information in existing methods, and improves the robustness and accuracy of graph representation learning.

The present application relates to the technical field of rotating machinery fault diagnosis and condition monitoring, and discloses a kind of blade crack fault feature mining and early warning method, comprising: obtaining blade simulation crack measured signal, source domain dataset is constructed, target domain dataset is constructed, and source domain dataset and target domain dataset form input tensor;Utilize one-dimensional deep residual network as feature extractor, and utilize source domain dataset to pre-train feature extractor, obtain pre-trained feature extractor;Unified framework containing pre-trained feature extractor, fault classifier and domain discriminator is constructed, and knowledge transfer from source domain to target domain is realized by joint optimization classification loss, domain adversarial loss and maximum mean difference, and the trained cross-domain transfer diagnosis model is obtained;Real-time monitoring signal is input into the trained cross-domain transfer diagnosis model, and fault probability vector is obtained, multi-level early warning is carried out in combination with adaptive threshold, which can provide technical support for unit safe service.